Luisa Giacomini

A switching controller for systems with hard uncertainties

A novel control scheme is proposed to solve the problem of stabilizing nonlinear uncertain systems with an unknown sign of the control gain and potentially unbounded nonlinearities. If the drift term f(x) has a growth rate such that the overall system has a finite escape time t/sup */, only a guess of a lowerbound, different from zero, of t/sup */ is necessary to design the control law. In spite of the hard uncertainties which affect the system model, the proposed switching controller is capable of asymptotically stabilizing the origin of the state space. The properties of the proposed control algorithm are compared with those of a well-known adaptive-control procedure when applied to the same class of systems.

Brief On modular backstepping design with second order sliding modes

A partially recursive backstepping-like procedure to design estimation-based controllers for uncertain nonlinear systems is presented in this paper. Its motivation relies on the intent of reducing the computational load of the backstepping design by exploiting the simplicity of sliding mode control. The stability features of the system controlled via the proposed approach are semi-global. Its transient behaviour turns out to be comparable with that obtained via the purely backstepping design. The proposed approach exhibits modularity, in the sense that the controller is proved to guarantee input-to-state stability regardless of the choice of the parameter estimation mechanism.

UML and Petri nets for design and analysis of distributed systems

This paper presents a modification to UML to improve the modelling and analysis of discrete-event dynamic system (DEDS) representations of manufacturing systems. It shows how Petri nets can be used to improve the representation and analysis of the dynamic model of a system specified using UML. Finally the technique is illustrated by its application to a simplified production line.

A backstepping second order variable structure control design for a class of uncertain nonlinear systems

The problem of the stabilization of uncertain nonlinear systems expressible in the perturbed-chain-of-integrators form is considered. Under suitable assumptions, the control problem is solved via a combined backstepping/variable structure control strategy which retains n-1 steps of a backstepping like procedure, and associates with them a final step aimed at generating a second order sliding regime on a suitably selected sliding manifold.

Design of Distributed Manufacturing Systems using UML and Petri Nets

Abstract This paper describes the design of a supervisory control system for a distributed manufacturing process, which forms part of a wider manufacturing system. The focus of the paper is on the design of a verifiable discrete event controller using a UML based method. The approach adopted involves (i) using Petri net models instead of conventional Statecharts to provide analytic Dynamic Models; and (ii) using compositional Petri net techniques to synthesise the Interconnection Model. The model of the complete controller can be then analysed and verified using Petri net theory. The approach is demonstrated by application to a prototype packaging machine.

Modular backstepping design of an estimation-based sliding mode controller for uncertain nonlinear plants

A partially recursive backstepping-like procedure to design estimation-based controllers for uncertain nonlinear systems is presented. Its motivation relies on the intent of reducing the computational load of the n-steps backstepping design by exploiting the promising features of second order sliding mode control. The proposed approach exhibits modularity, in the sense that the controller is proved to guarantee input-to-state stability regardless of the choice of the parameter estimation mechanism.

Control of nonholonomic systems with uncertainties via second order sliding modes

The problem of controlling a class of nonholonomic systems affected by uncertainties is addressed in this paper. The idea is to apply the so-called second order sliding mode control methodology. The crucial point is the necessity of reformulating the system model so as to make the uncertain terms appear in matched form. To this end, a suitable change of coordinates is performed and a peculiar sliding manifold is designed upon which a second order sliding mode is enforced. Apart from the robustness features, typical of the sliding mode approach, the proposed law has the advantage of being continuous, thus more acceptable in some applications such as those of mechanical nature

A simplified adaptive control scheme based on a combined backstepping/second order sliding mode algorithm

This paper presents a new combined adaptive/variable structure control approach for a class of nonlinear uncertain systems. In contrast to the classical backstepping procedure, it allows the presence of nonparametric uncertainties in the last two equations of the system. The advantages of the combined approach are mainly due to the enhancement of robustness and the reduction of the computational load. The stability and convergence features of the original adaptive scheme are retained.

A hybrid sliding mode control for systems with hard uncertainties

A control approach for continuous time uncertain nonlinear systems with unknown high frequency gain sign is presented, relying on a logic-driven sliding mode control. The resulting control strategy turns out to be of hybrid nature since the relevant features of the control action are modified on the occurrence of events tied to the state evolution.

Stabilization of Nonholonomic Uncertain Systems Via Adaptive Second Order Sliding Mode Control

Nonholonomic systems have been the object of research for control theorists for many years [24]. The key difficulty in controlling this class of systems, like wheeled mobile robots and wheeled vehicles [24], is tied to the fact that nonholonomic systems do not satisfy Brockett’s necessary smooth feedback stabilization condition [8] as shown in [29]. The control problem is further complicated whenever uncertainties of various nature affect the nonholonomic system model.